Skip to content

Work

Work

What the client asked for, what the problem turned out to be and what the work achieved.

Most of our work is confidential. Below are the cases we can publish and 11 engagements we can name. An owner who wants a reference gets one by phone.

Cases

Cases

Livesport coordinate Livesport's internal systems must stay correct under spiky load, because an error shows up as wrong scores for tens of millions of people within seconds. We designed, built and operated nine internal systems for Livesport in TypeScript and Node.js, and built them so Livesport's own team can maintain them. In production Nine internal systems running in production behind a platform with more than 100 million users. Flexiana (own product: Gitrevio) predict Owners could see commits, tickets and deploys, but learned what that work did to cost, risk and margin only after the books had closed. We built and run Gitrevio. It reads 31 engineering, project and HR sources, matches each person to one identity, forecasts what each project will cost to finish and flags margin and delivery risk while there's still time to act. In production A security-hardened product that runs inside Flexiana and is sold at gitrev.io. Flexiana Decision Lab (Rainbow) verify An allocation rule can pass every test and still break a capacity limit, a contract or an exclusivity clause on the one input no test tried. We built Rainbow, a theorem prover in Clojure with a small trusted kernel, and put it to work on our catalogue of operations-research algorithms. Formally proved Kernel-checked theorems about all 222 algorithms in our OR catalogue. Illustrative example (not a client engagement) select With more work on offer than people to do it, a business keeps choosing which orders to take, at what price and start date, with margin and penalties at stake. The model picks the orders with the best expected margin, marked down for the risk of a big loss, for shaky delivery and for leaning on one client. It never books more people or skills than you have. Backtested On synthetic data it kept more margin for the same risk of loss than a simple first-come, highest-price rule.

Selected engagements

Engagements we can name

They span sports data at 100M+ users, fintech and payments, mortgage, govtech, SaaS, video infrastructure, fashion and HR technology.

  • Information System · Software

    Livesport

    Built nine internal systems behind one of the world's most-visited sports-results platforms, with more than 100 million users.

  • Design · Software

    Stylitics

    Product design and software engineering.

  • Software

    Guaranteed Rate

    Software engineering.

  • Software

    Cloudpermit

    Software engineering.

  • Design · HRTech · Software

    Frankie

    HRTech product design and engineering.

  • HRTech

    Lifesize Video & Audio

    Video and audio infrastructure.

  • FinTech · Mobile

    Zimpler

    Fintech and mobile engineering.

  • Blockchain · Design · Software

    Cicada Wallet

    Blockchain wallet engineering and product design.

  • Software

    Gitrevio

    Our own product. It shows owners where their engineering money goes.

    Visit
  • SaaS · Software

    Crossbeam

    We built Crossbeam's partner-ecosystem platform alongside their team, through product-market fit and fast growth. Their Director of Engineering tells the story on video.

    Visit
  • FinTech · Software

    Starcity

    Pulled payments out of a monolith into their own service, and automated payouts from an external payment provider to clients' bank accounts.

Next step

Bring us a decision

Tell us about the choice or the operating problem. A partner will tell you how we'd approach it.